As of today, we have limited knowledge on the extent to which older Americans with epilepsy, especially typically disadvantaged minorities, receive antiepileptic drug treatment (AED) and follow-up care that is safe and effective. Epilepsy incidence peaks in older age as well as in childhood, and is higher in some minorities; e.g., it is 80% higher in African American (AA) compared to white Medicare beneficiaries. Appropriate treatment is crucial to optimize the chances of seizure control. Based on adult treatment effectiveness and safety data, the Quality Indicators for Epilepsy Treatment (QUIET) were developed in 2007 to assess quality of care. Previous literature suggests not all racial groups receive QUIET concordant care. In addition, poor AED adherence, also typically more common among minorities, limits seizure control. Improving care and adherence to AED treatment in all older adults with epilepsy has the potential to prevent avoidable seizures and also reduce medical costs. Our long-term goal is to inform and design interventions that improve the health care for older minorities with epilepsy and reduce health care costs associated with preventable events. The objective of this application is to examine the current quality of, and adherence to, AEDs and the current quality of care after recurrent seizures across racial groups of older Americans, and to identify opportunities for quality of care improvements and health care costs reductions as a first step toward the long term goal. The rationale for this research is that it is fundamental to inform the progression toward the national objective of living well with epilepsy and potentially reduce health care costs for the hundreds of thousands Medicare beneficiaries who suffer from this disorder. We plan to accomplish our objective by using Medicare administrative claims and a validated algorithm to identify epilepsy cases from these data, to pursue the following specific aims: 1) Assess quality of AED treatment across racial groups of Medicare beneficiaries with epilepsy. Our hypothesis is that, compared to white beneficiaries, typically disadvantaged minorities (e.g., AAs, Native Americans, Hispanics) are less likely to have QUIET concordant AED treatment and less likely to adhere to AEDs; 2) Determine the quality of care after seizure recurrence across race. We hypothesize that minority beneficiaries are less likely to have QUIET concordant care after recurrent seizures. Using the Behavioral Model of Access to Care, analyses for aims 1 and 2 will determine the contribution to racial disparities of, among others, current features of drug plans and geographic variation in care; and 3) determine if lower health care costs are associated with QUIET concordant care across racial groups. If H1-H3 are correct, we hypothesize that minority beneficiaries will be more likely to have high potentially preventable health care costs. This novel project will have a positive impact because it will provide much needed information to move the field toward improving epilepsy care, reducing disparities and potentially reducing costs for older adults.